Peptide-Decorated Liposomes Promote Arrest and Aggregation of

Apr 3, 2012 - Platelet-mimetic synthetic hemostats are highly attractive in transfusion medicine. To this end, past research reports have described pa...
0 downloads 7 Views 4MB Size
Download PDF
Article pubs.acs.org/Biomac

Peptide-Decorated Liposomes Promote Arrest and Aggregation of Activated Platelets under Flow on Vascular Injury Relevant Protein Surfaces in Vitro Madhumitha Ravikumar, Christa L. Modery, Timothy L. Wong, and Anirban Sen Gupta* Case Western Reserve University, Biomedical Engineering, Cleveland, Ohio 44106, United States S Supporting Information *

ABSTRACT: Platelet-mimetic synthetic hemostats are highly attractive in transfusion medicine. To this end, past research reports have described particles that either amplify platelet aggregation or mimic platelet adhesion. However, a construct design that effectively combines both functionalities has not been reported. Here we describe the design of a liposomal construct simultaneously surface-decorated with three peptides (a vWFbinding peptide (VBP), a collagen-binding peptide (CBP), and an active platelet clustering cyclic-RGD (cRGD) peptide), that can integrate platelet-mimetic dual hemostatic activities of adhesion and aggregation. We first demonstrate that surface-immobilized cRGD-liposomes are capable of aggregating activated platelets onto themselves. Subsequently, we demonstrate that hetero-multivalent liposomes bearing VBP, CBP, and cRGD, when introduced in flow with ∼20, 000 activated platelets per microliter, are capable of adhering to vWF/collagen surfaces and promoting the recruitment/aggregation of platelets onto themselves. We envision that optimizing this construct can lead to a highly refined synthetic hemostat design for potential application in transfusion medicine.

■

INTRODUCTION Transfusion of natural platelet suspensions and platelet-derived products is routinely used clinically for the treatment of various hematologic and oncologic bleeding disorders as well as in trauma or surgery-related heavy bleeding complications.1−4 However, these natural platelet products are highly expensive, have limited availability due to shortage of blood donors, have a very short shelf life (3−7 days), and can present a variety of contamination and biological risks.5,6 Hence, there is a significant clinical interest in synthetic constructs that exhibit hemostatically relevant platelet-mimetic functionalities while allowing easy scalable fabrication, long shelf life, and reduced biological risks. Research in developing such synthetic platelet mimics over the past two decades has resulted in several designs (Table 1).7−18 As evident from the Table, most of these designs have focused on creating particles that can amplify the aggregation of platelets (particles coated with fibrinogen or fibrinogen-derived RGD and H-12 peptides), whereas a few have focused on creating particles that exhibit platelet-mimetic adhesion under flow (e.g., particles coated with GPIbα or GPIa/IIa motifs). The designs that just amplify platelet aggregation can render only partial benefit in hemostasis at high particle doses, as demonstrated in several reports based on bleeding models in small animals.14−17 Additionally, these reports are not mutually comparable because some have utilized thrombocytopenic animals with subacute bleeding (e.g., tail-vein or ear-punch bleeding), whereas others have utilized normal platelet count © 2012 American Chemical Society

animals with acute (traumatic) bleeding (e.g., femoral artery laceration). Also, some reports have used particle concentration in numbers per kilogram of animal body weight or per unit injection volume (n per kg or n per mL), whereas some others have used particle concentration in total mass per injection volume or per kilogram of animal body weight (mg per mL or mg per kg). Therefore, although some hemostatic effect of the various designs has been demonstrated in all of these reports, whether this effect is actually therapeutically efficacious remains to be a topic of further research. We rationalize that the hemostatic benefit from particles capable of just amplifying platelet aggregation will be therapeutically relevant and efficient only in heavy bleeding conditions with normal platelet counts because, in such cases, a large number of activated platelets will already be available, adhering (and being recruited) at the injury site, whose aggregation can then be kinetically enhanced by particlemediated clustering. However, in bleeding complications that are subacute or involve reduced platelet numbers (e.g., thrombocytopenic or thrombesthenic conditions), particles that just amplify platelet aggregation may not be therapeutically effective in maintaining hemostasis because they may not find sufficient number of activated platelets already adhering or being recruited to the bleeding site. Instead, these particles may end up clustering the low number of available active platelets in Received: February 3, 2012 Revised: March 27, 2012 Published: April 3, 2012 1495

dx.doi.org/10.1021/bm300192t | Biomacromolecules 2012, 13, 1495−1502

Biomacromolecules

Article

Table 1. Various Design Approaches Carried out in Past Research Regarding the Development of Platelet-Mimetic Synthetic Hemostat Particles

and ethanol were purchased from Thermo Fisher Scientific (Pittsburgh, PA). Cholesterol, dimethyl sulfoxide (DMSO), and collagen were purchased from Sigma Aldrich (Saint Louis, MO). Fluorescently labeled monoclonal antibody, Alexa Fluor 647-antiCD62P (staining activated platelet P-selectin), was purchased from BioLegend (San Diego, CA). The lipids distearyl phosphatidyl choline (DSPC), distearyl phosphatidyl ethanolamine (DSPE), poly(ethylene glycol)-modified DSPE (DSPE-PEG2000), carboxy-poly(ethylene glycol)-modified DSPE (DSPE-PEG2000-COOH), and biotinylated poly(ethylene glycol)-modified DSPE (DSPE-PEG2000-Biotin) were purchased from Avanti Polar Lipids (Alabaster, AL). ClearOx and N-hydroxysuccinimide-modified fluorescein (NHS-fluorescein) were purchased from Invitrogen (Carlsbad, CA). Human vWF (FXIII free) was purchased from Hematologic Technologies (Essex Junction, VT). The PPFC system for dynamic flow studies was purchased from Glycotech (Gaithersburg, MD). The dimensions of the chamber were flow width = 1 cm and distance between plates = 0.00254 cm. The peptide sequences used were TRYLRIHPQSWVHQI (VBP), [GPO]7 (CBP), and cyclo-CNPRGDY(OEt)RC (cRGD). The VBP, CBP, and the linear precursor of the cRGD peptide were synthesized using fluorenylmethyloxycarbonyl chloride (FMoc)-based solid-phase chemistry on a Knorr resin and characterized using mass spectroscopy. The linear precursor of cRGD was subjected to sulfhydril oxidation of cysteine termini using ClearOx reagent to achieve disulfide-based cyclization. The rationale behind using these peptides has been explained in previous publications.20,21 Preparation of Platelet Suspensions. Venous blood from healthy, medication-free, adult donors was drawn into 3.8% w/v sodium citrate anticoagulant at a 9:1 ratio (by volume), in compliance with CWRU IRB-approved protocols. Platelet-rich plasma (PRP) was obtained by centrifuging human whole blood at 150 g for 15 min, and platelet count was monitored using a Coulter Counter. To prepare LPCs mimicking thrombocytopenic conditions, we further centrifuged a portion of PRP at 2500g for 25 min to obtain platelet-poor plasma (PPP). This PPP was then added volumetrically to PRP such that final platelet concentration was adjusted to ∼20 000/μL, as monitored by a Coulter Counter. These LPC suspensions were used immediately after preparation. Fabrication of Surface-Modified Liposomes. The VBP, CBP, and cRGD peptides were conjugated via their N-termini to the carboxyl terminus of DSPE-PEG2000-COOH via standard carbodiimide chemistry adapted from previously reported methods,21,22 resulting in the various DSPE-PEG-peptide molecules. NHS-fluorescein was reacted with DSPE-PEG2000-COOH to form DSPE-PEG-fluorescein for fluorescence labeling of liposomes. DSPE-PEG-peptides were mixed at specific mol % with DSPC, cholesterol, DSPE-PEG, DSPEPEG-biotin, and DSPE-PEG-fluorescein. These mixed lipid formulations were used toward fabricating liposomes using standard reverse-phase

circulation to form free-floating aggregates that may increase the risk of embolism. Similarly, for particles having only the functionalities that exhibit platelet-mimetic adhesion,10−14 there may be partial benefit if the particles attaching at the injury site can themselves staunch bleeding. However, this may not be therapeutically sufficient because the particles cannot further amplify active platelet aggregation at the site. Interestingly, in a recent publication by Okamura et al.19 it was demonstrated that synthetic particles surface-modified by “platelet aggregation” promoting motifs (fibrinogen γ-chain derived H-12 peptides) and those surface-modified by “vWF-adhesion” promoting motifs (recombinant rGPIbα fragments), when introduced together in a thrombocytopenic bleeding model in rabbits, showed significantly higher hemostatic efficacy compared with either of these particles injected alone. These results suggest that the simultaneous presence of platelet-mimetic “adhesion-promoting” and “aggregation-promoting” functionalities on a single particle may result in a more efficacious design of a synthetic hemostat. On the basis of the above rationale, here we describe the development and experimental results from integrating plateletmimetic adhesion- and aggregation-promoting functionalities on a single particle, by decorating the surface of 150 nm diameter liposomes simultaneously with three peptides: a vWFbinding peptide (VBP), a collagen-binding peptide (CBP), and an active platelet GPIIb-IIIa-binding peptide (cRGD). We have previously demonstrated that liposomes bearing VBP and CBP motifs undergo platelet-mimetic adhesion under flow on vWF and collagen-coated surfaces in vitro at low-to-high shear in parallel plate flow chamber (PPFC) experiments.20 Here we first demonstrate that cRGD-modified liposomes pre-adhered to a surface can enhance the aggregation of ADP-activated platelets onto themselves, even at a low platelet concentration (LPC). Subsequently, we demonstrate that liposomes bearing all three peptides (VBP, CBP, and cRGD), when introduced in a PPFC flow system along with low concentration of ADPactivated platelets over a vWF/collagen mixed coated surface, are able to adhere to the surface under high shear and promote arrest and aggregation of active platelets onto sites of liposome adhesion.

■

MATERIALS AND METHODS

Materials. Phosphate-buffered saline (PBS), 3.8% w/v sodium citrate, paraformaldehyde (PFA), avidin, bovine serum albumin (BSA), 1496

dx.doi.org/10.1021/bm300192t | Biomacromolecules 2012, 13, 1495−1502

Biomacromolecules

Article

Figure 1. Schematic of methods to fabricate functionally integrated liposomes bearing VBP and CBP peptides for platelet-mimetic vWF and collagen adhesion under flow and cRGD peptides to promote arrest and aggregation of active platelets via interaction with integrin GPIIb/IIIa.

Figure 2. Schematic representation and representative fluorescence images from studies of platelet aggregation in the absence or in the presence of ADP on unmodified versus cRGD-modified biotinylated liposomes, pre-adhered as a monolayer on avidin-coated coverslips. Only cRGD-modified liposomes show enhanced arrest and aggregation of platelets (red fluorescence) on them in the presence of ADP-induced activation. For imaging, platelets were stained with P-selectin specific Alexa Fluor 647-anti-CD62P antibody and imaged using a Zeiss Axio Observer.D1 inverted fluorescence microscope fitted with a photometrics chilled CCD camera and a 63× objective. These studies were carried out in a gyratory shaker under gentle agitation (100 rpm), which results in very low (almost static) wall shear stress between 0.5 and 1.25 dyn/cm2. evaporation and extrusion techniques, as previously described.23 Extrusions were carried out near the glass-transition temperature of DSPC (∼60 °C) through nanoporous (200 nm pore-size) polycarbonate membranes, resulting in unilamellar liposomal constructs (∼150 nm average diameter). A general schematic of fabricating the “functionally integrated” liposomes (simultaneously bearing all three peptides VBP, CBP and cRGD) is shown in Figure 1. In Vitro Platelet Aggregation Studies. For studying whether the cRGD-modified liposomal constructs pre-adhered to a surface can induce aggregation of activated platelets even from LPCs, DSPC (49 mol %), cholesterol (45 mol %), DSPE-PEG (2.5 or 5 mol %), and DSPE-PEG-Biotin (1 mol %,) was combined with or without DSPEPEG-cRGD (2.5 mol %) to form cRGD-modified or unmodified

biotinylated liposomal constructs. These nonfluorescent cRGDmodified or unmodified biotinalyated liposomes were incubated with avidin-coated glass coverslips for 1 h and subsequently washed with PBS to remove any loosely bound liposomes. This produced coverslips with a stable coating of cRGD-modified or unmodified liposomal constructs, as shown in the schematic of coverslips in the left columns of Figure 2. LPC was obtained as previously described and incubated with the construct-coated coverslips for 1 h in the absence or in the presence of platelet agonist ADP, under gentle agitation (100 rpm in gyratory shaker). Previous studies have shown that speeds of 50− 100 rpm in gyratory shakers result in very low wall shear stress between 0.5 and 1.25 dyn/cm2.24 Post-incubation, the coverslips were gently washed with PBS to remove loosely bound platelets from the 1497

dx.doi.org/10.1021/bm300192t | Biomacromolecules 2012, 13, 1495−1502

Biomacromolecules

Article

modified by all three peptides), over the coated glass slides. The flow was maintained to produce wall shear stresses of 5−55 dyn/cm2 for 30 min in a closed loop circulation. In brief, the wall shear stress (τw) of the PPFC can be modulated as a function of flow rate (Q) by: τw = 6 μQ/bh2, where μ = fluid viscosity (0.015 dyn/cm2), b = width of the chamber (1.0 cm), and h = distance between plates (0.00254 cm).26 After 30 min, flow of just PBS was maintained for an additional 15 min in an open loop to remove any loosely bound constructs and platelets. The working hypothesis for this experimental design was that liposomal constructs bearing all three peptides (VBP, CBP, and cRGD) will be able to adhere stably to the vWF/collagen surface under low-to-high shear flow, recruit activated platelets in flow, and promote aggregation of the activated platelets onto themselves at sites of liposome adhesion. Liposomal constructs bearing only “adhesive” peptides (only VBP and CBP) or only “aggregatory” peptide (only cRGD) will have much reduced capability of demonstrating plateletmimetic dual functions of promoting adhesion and arrest/aggregation of active platelets from flowing LPC suspensions. The slides were imaged at various time points (5, 15, 30, and 45 min) of flow using an inverted fluorescence microscope. For each image, liposome fluorescence (green) and activated platelet fluorescence (red) intensity were quantified using Adobe Photoshop CS4 software. The colocalization of these two fluorescence colors was considered to be a quantitative measure of liposomes adhering to the vWF/collagen surface under flow and then promoting arrest and aggregation of activated platelets onto themselves. This colocalization is qualitatively shown in pseudo-colored yellow overlay in the results. The colocalization was quantified using Axiovision software by acquiring the percentage of green pixels that also had red pixels superposed on them (at fixed pixel size) for every image and multiplying this percentage by the pixel-averaged green fluorescence intensity for that specific image. All statistical analyses were performed using ANOVA, and significance was considered at p < 0.05.

construct-coated surface. Subsequently, the coverslips were stained with mouse anti-human Alexa Fluor 647-anti-CD62P (red fluorescence, λmax ≈ 570 nm), which labels P-selectin on activated platelets. These stained coverslips were mounted onto glass slides, and the fluorescence of active platelets aggregated onto the coverslips was imaged using inverted fluorescence microscopy. The working hypothesis behind these experiments was that coverslips coated with cRGD-modified liposomal constructs would induce enhanced GPIIb-IIIa-binding mediated aggregation of ADP-activated platelets compared with the controls. In the absence of cRGD-modification on liposomes (unmodified liposome coating), a percentage of ADP-activated platelets may still undergo some clustering mediated by the fibrinogen present in the plasma of LPC suspension, but these platelet clusters will aggregate only minimally on the unmodified liposome surface because bare or PEG-ylated phospholipids (liposome membrane component) are known to prevent platelet adhesion and arrest.25 Platelet aggregation was quantified as the percentage of coverslip surface area covered by platelet fluorescence. All statistical analysis was performed using ANOVA, and significance was considered to be p < 0.05. In Vitro Evaluation of Functionally Integrated Liposomal Constructs. For developing functionally integrated liposomal constructs where the platelet-mimetic “matrix-adhesion” and “aggregation” properties are combined on a single-particle platform, DSPEPEG-VBP (1.25 mol %), DSPE-PEG-CBP (1.25 mol %), and DSPEPEG-cRGD (2.5 mol %) were combined with DSPC (49 mol %), cholesterol (45 mol %), and DSPE-PEG-fluorescein (1 mol %). Negative control liposomal constructs did not contain any lipid− peptide conjugate in their formulations but instead contained 5 mol % DSPE-PEG. Comparison liposomal formulations contained only “adhesion” functionality (1.25% DSPE-PEG-VBP and 1.25 mol % DSPE-PEG-CMP together with 2.5 mol % DSPE-PEG, 49 mol % DSPC, 45 mol % cholesterol and 1 mol % DSPE-PEG-fluorescein) or only “aggregatory” functionality (2.5 mol % DSPE-PEG-cRGD together with 2.5 mol % DSPE-PEG, 49 mol % DSPC, 45 mol % cholesterol, and 1 mol % DSPE-PEG-fluorescein). For these experiments, glass slides were coated with adjacent circular regions of albumin (control surface with no specific adhesive interaction with any liposome formulation) and 50:50 vWF/collagen (vascular injury site mimicking protein surface with adhesive interaction with VBP- and CBP-decorated liposomes). The coated glass slides were vacuum sealed within the PPFC chamber, with the coated sides exposed to the flow (schematic shown in Figure 3). Platelets in LPC were

■

RESULTS AND DISCUSSION Platelet Aggregation on Biotinylated cRGD-Modified Liposomes Coated on Avidin Surfaces. The last column in Figure 2 shows representative fluorescence images for platelet interaction with the coverslip-coated liposome formulations in the absence or in the presence of ADP. Figure 4 shows the corresponding quantitative data from these studies. The images and the data indicate that in the absence of ADP-induced activation, quiescent platelets hardly undergo any interaction with the liposome layer, irrespective of whether the liposomes were unmodified or cRGD-modified. This also suggests that the liposomes themselves, whether unmodified or cRGD-modified, do not activate (and hence aggregate) quiescent platelets. In contrast, upon ADP-induced activation, platelets undergo significantly enhanced interaction with the coverslip-adhered cRGD-modified liposomes, resulting in high extent of platelet aggregation. In comparison, the unmodified liposomes show only minimal aggregation of activated platelets onto themselves. This establishes that the cRGD-modified liposomal constructs adhered onto a surface are capable of promoting recruitment and aggregation of activated platelets onto themselves. Evaluation of Functionally Integrated Liposomal Constructs in Promoting Arrest and Aggregation of Platelets on vWF/Collagen Surface under Flow. Figure 5 shows a representative set of fluorescence images, and Figure 6 shows the quantitative data from the PPFC studies evaluating the various liposomal constructs interacting with activated platelets while flowing over the vWF/collagen-coated or the albumin-coated surface. In the images, the green fluorescence represents adhered liposomal constructs, the red fluorescence represents arrested and aggregated active platelets, and the yellow pseudo-color represents colocalization of the green

Figure 3. Schematic of the parallel plate flow chamber (PPFC) preparation with adjacent albumin-coated (negative control) and mixed (vWF/collagen)-coated areas and anticipated interaction of peptide (VBP and CBP)-modified liposomes with the surfaces under flow. preincubated with ADP and prestained with red fluorescent Alexa Fluor anti-CD62P. These LPC suspensions were allowed to flow through the PPFC along with various formulations of green fluorescent liposomes (unmodified, only adhesive peptide-modified, only aggregatory peptide-modified, or functionally integrated ones 1498

dx.doi.org/10.1021/bm300192t | Biomacromolecules 2012, 13, 1495−1502

Biomacromolecules

Article

to vWF/collagen surface (minimal green fluorescence) and consequently do not promote any arrest and aggregation of active platelets (minimal yellow fluorescence), as shown in the fourth row of images in Figure 5. As before, some platelet fluorescence (red) is seen here on the vWF/collagen surface due to direct interaction and arrest of activated platelets on this surface. In contrast with all of these, functionally integrated liposomal constructs bearing all three peptides (VBP, CBP, and cRGD) show high extent of green fluorescence as well as red fluorescence with significant yellow overlay, suggesting colocalization of the green fluorescent liposomes and the red fluorescent platelets on the vWF/collagen surface (third row of images in Figure 5). This indicates the enhanced ability of these functionally integrated constructs to undergo stable adhesion to the vWF/collagen surface under flow and promote arrest and aggregation of activated platelets onto themselves, mimicking the primary hemostatic action of natural platelets. The qualitative results indicated by the fluorescence images are further validated by the quantitative data analysis for liposome fluorescence, platelet fluorescence, and colocalized fluorescence intensity shown in separate graphs in Figure 6. From the graphs, it is evident that the functionally integrated liposomes have significantly enhanced ability to adhere to the vWF/collagen surface and promote arrest and aggregation of activated platelets onto them (blue bars in the colocalization graph) compared with unmodified, pro-aggregatory, or matrixadhesive liposomes. The pro-aggregatory liposomes (modified by cRGD only) seemed to cause statistically higher aggregation compared with the unmodified and the matrix-adhesive liposomes (red bars compared with the brown and green bars in the colocalization graph), but this is probably an effect of the cRGD-modified liposomes causing clustering of active platelets in free flow and some of the heavier clusters migrating down and sticking to the vWF/collagen surface. However, this effect of pro-aggregatory liposomes is still statistically lower than the action of the functionally integrated liposomal constructs. These results establish that combining the platelet-mimetic key hemostatic functionalities of adhesion-promotion and aggregation-promotion on a single particle platform can lead to a more refined design of a synthetic hemostat. In the native mechanism of platelet-mediated primary hemostasis in vascular injury, platelets initially adhere to vWF at the injury site via interaction between GPIbα of platelet surface receptor complex GPIb-IX-V. This adhesion is enhanced with increasing shear as vWF multimerizes with increasing shear, thereby allowing larger extent of GPIbα interaction. The GPIbα-vWF interaction is supplemented by additional binding interaction of platelet surface receptors GPVI and GPIa/IIa to fibrillar collagen, which secures the “rolling” vWF-adhered platelets and arrests them at the injury site. In our design, these two mechanisms of platelet adhesion are mimicked by decoration of multiple VBP and CBP copies on the liposome surface. Furthermore, in natural hemostasis, the arrested, adhered platelets get activated and act as nucleation points for recruitment and aggregation of more activated platelets via interaction between native ligand fibrinogen with the surface integrin GPIIb-IIIa on active platelets. In our design, to mimic and amplify this process, we decorated the liposome surface by multiple copies of fibrinogen-mimetic cRGD peptides, which have high affinity and selectivity to active platelet GPIIb-IIIa. The results from in vitro PPFC studies with “functionally integrated” liposomal constructs establish successful platelet-mimicry of our design. Although in our experiments the

Figure 4. Quantitative analysis of platelet aggregation studies described in Figure 2. Results show percent of surface area covered in red fluorescence (aggregation of platelets to cRGD-modified or unmodified liposome surfaces in the presence and in the absence of ADP under gentle agitation at 100 rpm in a gyratory shaker). cRGDmodified liposomes promote significant aggregation of ADP-activated platelets compared with the other conditions (p < 0.05).

liposomes and red fluorescent platelets in the same field of view. As evident from the images in the fifth row of Figure 5, the albumin surface showed hardly any adhesion of liposomes or arrest of platelets, and consequently, the quantitative values of liposome or platelet fluorescence (and colocalization) from the albumin surfaces are not included in the quantitative data in Figure 6. Although images were taken at six shear values between 5 and 55 dyn/cm2 and at four time points between (5, 15, 30, and 45 min), representative fluorescence images are shown at only one shear value (35 dyn/cm2) and one final time point (45 min) for convenience. The quantitative data in Figure 6 are shown for this 45 min time point across all shear stress values studied. From the images in Figure 5, it is evident that liposomes bearing only aggregation-promoting cRGD peptides are unable to undergo significant adhesion to the vWF/collagen surface and promote aggregation of activated platelets from the LPC condition onto themselves, even if they may cluster some activated platelets in flow. This is suggested by the minimal green fluorescence and minimal yellow colocalization shown in the second row of images. The red fluorescence shown in this row indicates a certain extent of platelet arrest and aggregation, which is possibly due to the direct interaction of activated platelets with the vWF/collagen surface rather than being mediated by liposomes. Constructs bearing only adhesionpromoting peptides (VBP and CBP) but no cRGD can adhere stably to vWF/collagen surface under flow but are unable to promote significant arrest and aggregation of activated platelets onto themselves from the flowing LPC suspension. This is suggested by the presence of considerable green fluorescence but minimal yellow colocalization in the first row of images. As before, the red fluorescence shown in this row indicates a certain extent of platelet arrest and aggregation due to the direct interaction of activated platelets with the vWF/collagen surface rather than being mediated by liposomes. Unmodified liposomes (no peptide modification) show insignificant adhesion 1499

dx.doi.org/10.1021/bm300192t | Biomacromolecules 2012, 13, 1495−1502

Biomacromolecules

Article

Figure 5. Experimental setup and resulting representative fluorescence images from studies involving unmodified and various peptide-modified liposomal constructs under flow with low concentrations of activated platelets through the PPFC over albumin-coated and vWF/collagen-coated surfaces. Green fluorescence (from DSPE-PEG-Fluorescein) represents liposomal constructs adhered onto vWF/collagen and albumin surfaces. Red fluorescence (from Alexa Fluor 647 -anti-CD62P) represents activated platelets aggregated onto vWF/collagen and albumin surfaces in the same field of view as the liposome adhesion images. Yellow overlay represents colocalization of the green and red fluorescence signifying adhered liposome-promoted aggregation of activated platelets. Only images at the 45 min time point for shear stress value of 35 dyn/cm2 are shown here for convenience.

various peptides were presented on a liposome surface, they can be potentially conjugated to any other particle platform if needed. Also, in the experiments reported here, the absolute mol % of the different peptides in the various liposome formulations was kept constant. (VBP and CBP were always at 1.25 mol % each, and cRGD was always at 2.5 mol %.) For “adhesion only” liposomes (VBP and CBP modification only) and “aggregation only” liposokmes (cRGD modification only), 2.5 mol % of just DSPE-PEG was added along with the DSPE-PEGpeptide conjugates to keep the total PEG-ylation content at 5 mol %. In the “functionally integrated” liposomes, this was achieved by combining VBP and CBP at 1.25 mol % each and cRGD at 2.5 mol %, and hence no additional DSPE-PEG was incorporated in these formulations. This initial metric of peptide composition was chosen to investigate the feasibility of our platelet-mimetic design approach. Our results successfully demonstrate the platelet-mimetic bioactivity of our design as a function of the “type” of peptide incorporation. It can be expected that along with the type, the relative densities of peptide incorporation will also influence the platelet-mimetic bioactivities of such constructs. Our ongoing and future studies are focused on modulating the peptide incorporation percentages to optimize their surface decoration density for the best

hemostatic performance, and this will be an area covered in future publications. Besides transfusion of natural platelet-based products, there are a few other synthetic or recombinant hemostatic products in current clinical use for treatment of bleeding complications such as QuikClot (a zeolite that induces platelet aggregation)27,28 and NovoSeven (coagulation factor VII that triggers the activation of the coagulation cascade).29,30 QuikClot is relevant only in topical application on open wounds and not in intravenous transfusion medicine. However many bleeding complications are internal, where hemostasis will require intravenous therapy. NovoSeven can be used intravenously but has been reported to present risks of immunogenic and thromboembolic complications.31,32 In addition, being a coagulation factor, it requires the presence of other factors and activated platelet membranes for effective action. Therefore, the application and efficacy of these products in hemostasis are limited. In contrast, a synthetic particle construct that avoids biologic contamination risks and side effects, has convenient preparation methodology, and has a long storage life can be potentially administered intravenously, can efficiently mimic both adhesive and aggregatory actions of natural platelets toward primary hemostasis, and can have significant potential in transfusion medicine. Our biomimetic approach of 1500

dx.doi.org/10.1021/bm300192t | Biomacromolecules 2012, 13, 1495−1502

Biomacromolecules

■

Article

CONCLUSIONS We have successfully demonstrated that liposomes heteromultivalently surface-modified with peptides mimicking platelet’s adhesion and aggregation properties can adhere to vascular-injury-relevant protein-coated surfaces and promote arrest and aggregation of active platelets onto themselves under a dynamic shear flow environment in vitro. The plateletmimetic activity of these liposomal constructs was demonstrated with LPCs, suggesting that in normal physiological platelet concentrations these constructs can potentially have even higher activity. Optimization of this approach can lead to a highly efficient design of synthetic hemostat with application in transfusion medicine.

■

ASSOCIATED CONTENT

S Supporting Information *

MALDI-TOF mass spectra for the peptides used to conjugate to the liposome surface to render platelet-mimetic pro-adhesive and pro-aggregatory properties. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The research was partially supported by American Heart Association Beginning-Grant-in-Aid no. 0765371B and Case Coulter Translational Research Partnership Pilot Funding. We would also like to acknowledge the facilities provided by the Center for Cardiovascular Biomaterials (CCB) in the department of Biomedical Engineering at Case Western Reserve University. The authors declare no competing financial interests.

■

REFERENCES

(1) Mohanty, D. Current concepts in platelet transfusion. Asian J. Transfus. Sci. 2009, 3 (1), 18−21. (2) Heal, J. M.; Blumberg, N. Optimizing platelet transfusion therapy. Blood Rev. 2004, 18 (3), 149−165. (3) Stroncek, D. F.; Rebulla, P. Platelet transfusions. Lancet 2007, 370 (9585), 427−438. (4) Pisciotto, P. T.; Benson, K.; Hume, H.; Glassman, A. B.; Oberman, H.; Popovsky, M.; Hines, D.; Anderson, K. Prophylactic versus therapeutic platelet transfusion practices in hematology and/or oncology patients. Transfusion 1995, 35 (6), 498−502. (5) Lee, D. H.; Blajchman, M. A. Platelet substitutes and novel platelet products. Expert Opin. Invest. Drugs 2000, 9 (3), 457−469. (6) Lee, D. H.; Blajchman, M. A. Novel treatment modalities: new platelet preparations and substitutes. Br. J. Haematol. 2001, 114 (3), 496−505. (7) Baldassare, J. J.; Kahn, R. A.; Knipp, M. A.; Newman, P. J. Reconstitution of Platelet Proteins into Phospholipid-Vesicles Functional Proteoliposomes. J. Clin. Invest. 1985, 75 (1), 35−39. (8) Coller, B. S.; Springer, K. T.; Beer, J. H.; Mohandas, N.; Scudder, L. E.; Norton, K. J.; West, S. M. Thromboerythrocytes. In vitro studies of a potential autologous, semi-artificial alternative to platelet transfusions. Journal of Clinical Investigation 1992, 89 (2), 546−555. (9) Rybak, M. E.; Renzulli, L. A. A liposome based platelet substitute, the plateletsome, with hemostatic efficacy. Biomater., Artif. Cells, Immobilization Biotechnol. 1993, 21 (2), 101−118. (10) Takeoka, S.; Teramura, Y.; Ohkawa, H.; Ikeda, Y.; Tsuchida, E. Conjugation of von Willebrand factor-binding domain of platelet glycoprotein Ib alpha to size-controlled albumin microspheres. Biomacromolecules 2000, 1 (2), 290−295.

Figure 6. Quantitative analysis of the fluorescence data showing (A) liposome adhesion (quantification of green fluorescence intensity), (B) platelet aggregation (quantification of red fluorescence intensity), and (C) colocalization of green and red fluorescence for PPFC experiments with unmodified and various peptide-modified liposomal constructs under flow with low concentrations of activated platelets through the PPFC over albumin-coated and vWF/collagen-coated surfaces. The data are shown for wall shear stress values of 5−55 dyn/ cm2 at the 45 min time point. The results demonstrate that the functionally integrated liposomes (surface-modified simultaneously with VBP, CBP, and cRGD) have significantly enhanced capacity to adhere to the vWF/collagen surface under flow at the various shear stress values and promote significant arrest and aggregation of active platelets onto themselves (p < 0.05).

combining the key hemostatic functions of platelets on a single particle addresses the above criteria and can therefore lead to a highly efficacious design of synthetic hemostats. 1501

dx.doi.org/10.1021/bm300192t | Biomacromolecules 2012, 13, 1495−1502

Biomacromolecules

Article

(11) Nishiya, T.; Murata, M.; Handa, M.; Ikeda, Y. Targeting of liposomes carrying recombinant fragments of platelet membrane glycoprotein Ibalpha to immobilized von Willebrand factor under flow conditions. Biochem. Biophys. Res. Commun. 2000, 270 (3), 755−760. (12) Davies, A. R.; Judge, H. M.; May, J. A.; Glenn, J. R.; Heptinstall, S. Interactions of platelets with Synthocytes, a novel platelet substitute. Platelets 2002, 13 (4), 197−205. (13) Nishiya, T.; Kainoh, M.; Murata, M.; Handa, M.; Ikeda, Y. Reconstitution of adhesive properties of human platelets in liposomes carrying both recombinant glycoproteins Ia/IIa and Ib alpha under flow conditions: specific synergy of receptor-ligand interactions. Blood 2002, 100 (1), 136−142. (14) Teramura, Y.; Okamura, Y.; Takeoka, S.; Tsuchiyama, H.; Narumi, H.; Kainoh, M.; Handa, M.; Ikeda, Y.; Tsuchida, E. Hemostatic effects of polymerized albumin particles bearing rGPIa/ IIa in thrombocytopenic mice. Biochem. Biophys. Res. Commun. 2003, 306 (1), 256−260. (15) Levi, M.; Friederich, P. W.; Middleton, S.; de Groot, P. G.; Wu, Y. P.; Harris, R.; Biemond, B. J.; Heijnen, H. F.; Levin, J.; ten Cate, J. W. Fibrinogen-coated albumin microcapsules reduce bleeding in severely thrombocytopenic rabbits. Nat. Med. 1999, 5 (1), 107−111. (16) Okamura, Y.; Fujie, T.; Nogawa, M.; Maruyama, H.; Handa, M.; Ikeda, Y.; Takeoka, S. Haemostatic effects of polymerized albumin particles carrying fibrinogen gamma-chain dodecapeptide as platelet substitutes in severely thrombocytopenic rabbits. Transfus. Med. 2008, 18 (3), 158−166. (17) Bertram, J. P.; Williams, C. A.; Robinson, R.; Segal, S. S.; Flynn, N. T.; Lavik, E. B. Intravenous hemostat: nanotechnology to halt bleeding. Sci. Transl. Med. 2009, 1 (11), 11ra22. (18) Okamura, Y.; Takeoka, S.; Eto, K.; Maekawa, I.; Fujie, T.; Maruyama, H.; Ikeda, Y.; Handa, M. Development of fibrinogen gamma-chain peptide-coated, adenosine diphosphate-encapsulated liposomes as a synthetic platelet substitute. J. Thromb. Haemostasis 2009, 7 (3), 470−477. (19) Okamura, Y.; Handa, M.; Suzuki, H.; Ikeda, Y.; Takeoka, S. New strategy of platelet substitutes for enhancing platelet aggregation at high shear rates: cooperative effects of a mixed system of fibrinogen gamma-chain dodecapeptide- or glycoprotein Ibalpha-conjugated latex beads under flow conditions. J. Artif. Organs 2006, 9 (4), 251−258. (20) Ravikumar, M.; Modery, C.; Wong, T.; Sen Gupta, A. Mimiking adhesive functionalities of blood platelets using ligand-modified liposomes. Bioconjugate Chem. 2012, underreview. (21) Huang, G.; Zhou, Z.; Srinivasan, R.; Penn, M. S.; KottkeMarchant, K.; Marchant, R. E.; Gupta, A. S. Affinity manipulation of surface-conjugated RGD peptide to modulate binding of liposomes to activated platelets. Biomaterials 2008, 29 (11), 1676−1685. (22) Simpura, I. Method for Synthesis of Phospholipids-PEGbiomolecule Conjugates. Eur. Pat. Appl. EP 1738770 A1, Jan 3, 2007. (23) Szoka, F.; Olson, F.; Heath, T.; Vail, W.; Mayhew, E.; Papahadjopoulos, D. Preparation of unilamellar liposomes of intermediate size (0.1−0.2 mumol) by a combination of reverse phase evaporation and extrusion through polycarbonate membranes. Biochim. Biophys. Acta 1980, 601 (3), 559−571. (24) Ley, K.; Lundgren, E.; Berger, E.; Arfors, K. E. Shear-dependent inhibition of granulocyte adhesion to cultured endothelium by dextran sulfate. Blood 1989, 73 (5), 1324−1330. (25) Ishihara, K.; Oshida, H.; Endo, Y.; Watanabe, A.; Ueda, T.; Nakabayashi, N. Effects of phospholipid adsorption on nonthrombogenicity of polymer with phospholipid polar group. J. Biomed. Mater. Res. 1993, 27 (10), 1309−1314. (26) Bacabac, R. G.; Smit, T. H.; Cowin, S. C.; Van Loon, J. J. W. A.; Nieuwstadt, F. T. M.; Heethaar, R.; Klein-Nulend, J. Dynamic shear stress in parallel-plate flow chambers. J. Biomech. 2005, 38 (1), 159− 167. (27) Robinson, K. Controlling bleeding in the field: hemostatic powders and dressings debut in the prehospital setting. J. Emerg. Nurs. 2004, 30 (2), 160−161. (28) Rhee, P.; Brown, C.; Martin, M.; Salim, A.; Plurad, D.; Green, D.; Chambers, L.; Demetriades, D.; Velmahos, G.; Alam, H. QuikClot

use in trauma for hemorrhage control: Case series of 103 documented uses. J. Trauma: Inj., Infect., Crit. Care 2008, 64 (4), 1093−1099. (29) Hedner, U. Recombinant factor VIIa (NovoSeven (R)) as a hemostatic agent. Semin. Hematol. 2001, 38 (4), 43−47. (30) Mayo, A.; Misgav, M.; Kluger, Y.; Geenberg, R.; Pauzner, D.; Klausner, J.; Ben-Tal, O. Recombinant activated factor VII (NovoSeven((TM))): addition to replacement therapy in acute, uncontrolled and life-threatening bleeding. Vox Sang. 2004, 87 (1), 34−40. (31) Felfernig, M.; Huepfl, M. Experience of recombinant activated factor VII (NovoSeven) in the operating theatre and intensive care unit for the management of intracranial bleeding in nonhaemophilic patients. Clin. Neurol. Neurosurg. 2008, 110 (3), 227−232. (32) Poon, M. C.; d’Oiron, R. Recombinant activated factor VII (NovoSeven) treatment of platelet-related bleeding disorders. International registry on recombinant factor VIIa and congenital platelet disorders group. Blood Coagulation Fibrinolysis 2000, 11 (Suppl 1), S55−S68.

1502

dx.doi.org/10.1021/bm300192t | Biomacromolecules 2012, 13, 1495−1502